Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already cover safety (readOnlyHint, idempotentHint, non-destructive, openWorld), yet the description adds substantive operational context they don't carry: cost is 1 credit (2 for xlsx output), output is returned inline capped at 1 MiB, input is up to 10 MB, and oversized inputs are rejected rather than silently truncated. These are exactly the cost/limit disclosures this dimension rewards.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Purpose and formats are front-loaded and the cost line is placed early, which is the right priority. However, the output portion is choppy ('Output: a file (...) returned inline, at most 1 MiB. The converted table. meta: {...}'), with a dangling fragment and a raw meta shape that reads as schema leakage rather than prose.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With no output schema, the description carries the return burden and mostly does: it names the media types returned inline, the 1 MiB cap, and the meta object fields (rows, cols, from, to, sheet/sheets, encoding, delimiter). For a 9-parameter tool this is close to complete, though error cases beyond the 413 rejection and multi-sheet output behavior are left implicit.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so every parameter including to, from, sheet, header, encoding, delimiter, max_rows and infer_types is already documented in the schema. The description only echoes a few of these ('format, delimiter and encoding detection, header control') without adding syntax or semantics beyond the schema, so the baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Convert tables between CSV, TSV, XLSX and JSON') and enumerates the exact formats handled. This cleanly separates it from siblings like csv_profile (profiling) and json_yaml_convert (structured-document conversion), so an agent can route without opening any schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies when to use it (any format-to-format table conversion) and notes 'Omit to auto-detect' for the source format, but never names an alternative tool or states an exclusion. No guidance on when csv_profile or json_yaml_convert is the better choice, so usage must be inferred from the format list.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.